The continuous growth in air transportation comes with undesirable environmental impacts. Aircraft noise pollution has emerged as one of the most concerning problems in recent years, both for airlines and airport development. In addition to disrupting the daily activities of residents living near airports, aircraft noise can negatively impact their health, communication, and sleep.

The current state-of-the-art environmental analysis tools required the simulation the simulation of all flight missions in order to generate cumulative noise contours and assess the noise impact of air traffic in a given regions over a given period of time. Moreover, this requirement to input data from all flight missions imposes high computational costs on policy maker who are analyzing and comparing...[ Read more ]

The continuous growth in air transportation comes with undesirable environmental impacts. Aircraft noise pollution has emerged as one of the most concerning problems in recent years, both for airlines and airport development. In addition to disrupting the daily activities of residents living near airports, aircraft noise can negatively impact their health, communication, and sleep.

The current state-of-the-art environmental analysis tools required the simulation the simulation of all flight missions in order to generate cumulative noise contours and assess the noise impact of air traffic in a given regions over a given period of time. Moreover, this requirement to input data from all flight missions imposes high computational costs on policy maker who are analyzing and comparing multiple possible scenarios. This approach could be computationally prohibitive.

In the proposed work, I aim to reduce this computational effort by eliminating the need to simulate each flight trajectory one by one. Instead, I derive a distribution function to represent the thousands of flight missions to be evaluated. The probabilistic approach is employed to generate the noise contours based on this distribution information. All flight path data used in this study was obtained from flights departing from and arriving at the Hong Kong International Airport (HKIA).

I start by employing the hierarchical density based spatial clustering of applications with noise (HDBSCAN) algorithm to cluster the flight paths. I then get the distribution of each point inside the clusters by using a histogram and then calculate the average latitude, longitude, altitude, and speed for each bin from the histogram. The results of this study are validated against the noise contours generated from the brute-force approach, and the computational cost reduction is quantified. The discrepancy between the brute-force approach and the probabilistic approach is found to be 0.12%. Using the proposed probabilistic approach yields a computational cost reduction of up to 85% on a 100 x 100 grid. The derived parametric representation of flight trajectories has potential for use in other applications as well, such as trajectory optimization and aircraft design problems.